KOI-4878.01: Kepler’s Most Earth-like Planet Candidate?

Launched on March 7, 2009, the objective of NASA’s Kepler mission was to observe the brightness of 150,000 stars in a 115 square degree patch of sky in the constellations of Cygnus and Lyra looking for periodic dips indicative of transiting exoplanets. The goal was to create a large database of exoplanets allowing scientists to explore the structure and diversity of planetary systems. Kepler’s design and observing strategy was tailored to allow the detection of Earth-size planets in Earth-like orbits around Sun-like stars with an apparent magnitude as dim as 12.

Diagram showing the major components of NASA’s Kepler spacecraft. (NASA/Kepler Mission/Ball Aerospace)

Before the start of the Kepler mission, it was hoped that a handful of potential Earth-analogs could be detected. While Kepler discovered 2,345 (and still counting) confirmed exoplanets during its primary mission running from May 2009 to May 2013, none of them were anything that could be considered a true Earth analog. The overwhelming majority of the finds from Kepler’s primary mission were larger planets in short period orbits around stars smaller than the Sun – the types of worlds which are not only much more common than Earth-like planets, but easier to detect. Probably the most Earth-like finds of the mission were Kepler 186f (see the Kepler 186 page) and Kepler 442b (see the Kepler 442 page). Unfortunately, these worlds are somewhat larger than the Earth, are slow or synchronous rotators and orbit dim M and K type stars, respectively. While they might be habitable, they would not be very Earth-like.

This plot shows the confirmed planet and planet candidates found as of June 2017 by NASA’s Kepler mission as a function of size (denoted by the size of the planet symbol), the energy they receive and their host star’s temperature. The green bands indicate various definitions of the habitable zone (HZ). Click on image to enlarge. (NASA/Ames Research Center/Wendy Stenzel)

But in addition to its confirmed exoplanet discoveries, Kepler also identified 2,420 exoplanet candidates during its primary mission which still require follow up observations to confirm their planetary nature. Among these are some more Earth-like planets orbiting Sun-like stars. One of these candidates which seems to get some noticed periodically on the web is KOI-4878.01 – the first exoplanet candidate detected orbiting Kepler Object of Interest 4878. So, what is the nature of this exoplanet candidate and its status?

Detection by Kepler

KOI-4878 is a V magnitude 12.4 star in the constellation Draco near the outer edge of the star field Kepler monitored during its primary mission. Because of its dimness, it just missed being cataloged in the large star surveys of the late 19^th and early 20^th centuries and, because of its minuscule proper motion, it never got the attention of the astronomers looking for dim stars with high proper motion during the 20^th century. KOI-4878 was first cataloged as part of the 2MASS (2 Micron All Sky Survey) project, which was performed from 1997 to 2001, receiving the designation 2MASS J19045474+5000488.

This is a Digital Sky Survey image centered on KOI-4878. The bright star at the bottom left edge of this 10 arc min wide image is HD 178208 with a V magnitude of 6.43. (DSSS via CDS)

With the selection of the star field Kepler was to monitor, KOI-4878 was subsequently included in the Kepler Input Catalog (KIC) receiving the designation of KIC-11804437. Using the best photometry available at the time, KOI-4878 was estimated to have an effective temperature of 6031 +143/-168 K – just a touch hotter than the Sun and consistent with being a late F-type star. Its radius was estimated to be 1.068 +0.389/-0.145 times that of the Sun with a mass of 0.972 +0.143/-0.110 that of the Sun. These values lead to an estimated luminosity of about 1.35 times that of the Sun and a distance of around 1,200 light years.

This plot shows the photometric measurements made of KOI-4878 during Kepler’s primary mission. The time of the transits of KOI-4878.01 are indicated by small arrows near the bottom of the plot. Click on image to enlarge. (NASA/Kepler Mission)

Analysis of Kepler’s primary mission data from its first quarter (Q1) to Q12 for KOI-4878 revealed three possible transit events equally spaced in time. With events observed during Q1, Q6 and Q11, a predicted fourth event was missed during Q16 because of periodic changes in Kepler’s attitude which brought the star into the field of view of an inoperable focal plane array. A post-Q16 analysis of the available data showed that the events occurred with a period of 449.015±0.021 days, had a duration of 12.6±1.0 hours and a depth of 94.2±13.1 ppm. The candidate planet was designated KOI-4878.01.

Here is a plot of Kepler photometric measurements phase folded to the proposed orbital period of KOI-4878.01 The small dots are individual measurements, the larger blue circles are binned values and the red line is the best fit transit model. Click on image to enlarge. (NASA/Kepler Mission)

Combining these Kepler analysis results with the KIC estimates of the host star’s properties, it is possible to derive the properties of KOI-4878.01. Assuming the orbital period is equal to the time between transits, the semimajor axis of the orbit of KOI-4878.01 would be 1.137 +0.053/-0.040 AU (with the uncertainty driven by the uncertainty in the host star’s mass). Combining this value with the estimated luminosity of KOI-4878, the effective stellar flux of this planet candidate (S_eff – the amount of energy received from the host star compared to what Earth receives from the Sun) is a very Earth-like 1.05. Based on the depth of the transit and the estimated size of KOI-4878, KOI-4878.01 would have a radius 1.04 +0.38/-0.14 times that of the Earth.

But are these properties consistent with KOI-4878.01 being potentially habitable? Based on observations of like-size exoplanets, the odds favor this world being a rocky planet like the Earth instead of a volatile-rich mini-Neptune with no prospects of being habitable (see “Habitable Planet Reality Check: Terrestrial Planet Size Limit”). The Earth-like S_eff value also looks promising. According to modeling by Kopparapu et al., the conservative inner edge of the habitable zone (HZ) for an Earth-size planet orbiting a Sun-like star is at a S_eff value of about 1.11 which corresponds to a distance of about 1.11 AU from KOI-4878. Any closer and a runaway greenhouse effect will set in along with permanent water loss. Even with the uncertainties in the properties of KOI-4878.01 and its host star, the odds favor that this planet candidate orbits inside the HZ with potentially very Earth-like conditions.

So, what is the holdup confirming this Earth-like planet candidate? As part of Kepler’s normal data reduction pipeline, KOI-4878.01 was subjected to a battery of tests to validate the find. The obvious sources of false positives like a distant variable star blended with the image of KOI-4878 have been confidently eliminated. The problem comes from some of the more subtle statistical tests. In particular, the way the centroid of the star’s image appears to shift during a transit is much larger than usual and suggests that the transit-like events might be an instrument artifact instead. What is needed are more observations not only of additional transits, but of KOI-4878 itself to get better values for its properties which in turn will help us get a better handle on the properties of the planet candidate.

New Observations

As a result of the need to better characterize all of the stars observed during Kepler’s primary mission, a number of teams of astronomers from around the globe have been coordinating their efforts to gather the needed data. One of these has been the LAMOST-Kepler project. LAMOST (Large Sky Area Multi-Object Fibre Spectroscopic Telescope) is a 4-meter-class meridian reflecting Schmidt telescope located at the Xinglong observatory in China’s Hebei Province. Able to simultaneously acquire spectra for up to 4,000 objects at a time using a system of optical fibers to feed a set of spectrometers, LAMOST is ideal for performing large-area spectroscopic surveys.

The LAMOST telescope at China’s Xinglong observatory was used to secure spectra of the stars Kepler observed including KOI-4878. (LAMOST)

Over the course of 38 nights between May 30, 2011 and September 29, 2014, a total of 14 star fields were observed in support of the LAMOST-Kepler project to cover the area observed by Kepler during its primary mission. The spectra were then analyzed to derive updated stellar properties. Among the 51,385 stars observed was KOI-4878. According the analysis presented in Frasca et al., KOI-4878 is a spectral type G4V star with an effective temperature of 5780±202 K – essentially identical to the Sun.

The next important piece of data needed to pin down the properties of KOI-4878 is its distance. ESA’s ongoing Gaia astrometric mission measured the parallax of KOI-4878 for the first time. The DR2 parallax results in a distance of 1,135±7 light years – just a touch closer than the KIC photometry-based estimate. The proper motion of KOI-4878 came in at tiny 6.24 milliarc seconds per year. This is a small fraction of the proper motion that would have brought this object to the attention of astronomers during the last century.

ESA’s ongoing Gaia mission is providing much needed information on the motions and distances to stars including KOI-4848. (ESA)

Combining these updated observations with existing photometry, it is possible to calculate better estimates for the properties of KOI-4878. With the distance of KOI-4878 being just a bit smaller than thought earlier, the luminosity of the star comes in at a slightly lower 1.16±0.03 times that of the Sun. Combining this with the new effective temperature from Frasca et al., the radius of KOI-4878 is now estimated to be 1.08±0.02 times that of the Sun. Scaling the KIC mass value based on this lower luminosity, I estimate the mass to be around 0.94 times that of the Sun. These updated values are consistent with KOI-4878 being somewhat more evolved (i.e. older) than the Sun. The star’s low metallicity and fairly high space velocity also suggest that KOI-4878 is older than the Sun.

With these updated properties for KOI-4878, it is possible to derive a fresh set of properties for KOI-4878.01. These new values, along with the earlier values based on the KIC data, are summarized in the table below.

Properties of KOI-4878.01

While the new radius of KOI-4878.01 is essentially identical to the earlier estimate and the new semimajor axis value is well within the uncertainty of the earlier value, the S_eff value has dropped somewhat placing the planet candidate more securely into the HZ of this system. The largest source of uncertainty in my calculation here is the mass of KOI-4878. It is possible that this star has a mass closer to that of the Sun’s than the one I estimated here. But in that case, the size of the semimajor axis of KOI-4878.01 would be slightly larger dropping the S_eff to 0.90 which is still deeper into this more conservative definition of the HZ. Based on these initial rough calculations, the case for KOI-4878.01 being an Earth-twin appears slightly better than before, if the candidate planet can be confirmed.

What’s Next?

While future observations of KOI-4878 will help astronomers further refine its properties, more observations of transits would help confirm the planetary nature of KOI-4878.01. Unfortunately, with a brightness change of just 94 ppm during a 13-hour transit, ground observatories will be unable to get the needed data with sufficient accuracy. NASA’s Hubble or Spitzer Space Telescopes could potentially get the required data, but it will be a tough sell for anyone to make a case that KOI-4878, out of the thousands of Kepler planet candidates, should be observed using these very busy astronomical assets.

NASA’s TESS (Transiting Exoplanet Survey Satellite) is currently performing a survey for transiting exoplanets. (NASA)

NASA’s TESS (Transit Exoplanet Survey Satellite) satellite, which was launched on April 18, 2018 and is currently in the second year of its survey, could offer some help. TESS uses a set of four cameras to observe a “sector” running from near the ecliptic up to the ecliptic pole monitoring the thousands of stars in the field of view for periodic dips in brightness indicative of an orbiting planet. After four weeks, TESS changes its pointing to an adjacent sector with an ecliptic latitude about 27° to the west until virtually the whole sky is observed after a nominal two-year mission.

This all-sky map (in ecliptic coordinates) shows the first 21 sectors to be surveyed by TESS. The approximate position of KOI-4878 is indicated by the star and should be observed in Sector 26 when the next transit of KOI-4878.01 is predicted for June 18, 2020. Click on image to enlarge. (NASA/TESS)

Unfortunately, a predicted transit of KOI-4878.01 on March 27, 2019 was missed because TESS was still surveying the southern sky. The next transit, which is predicted to happen on June 18, 2020, appears to be quite promising, however. While the officially posted plans for future TESS observations currently run up only to Sector 21 scheduled to be observed starting on January 21, 2020, based on extrapolations it appears likely that KOI-4878 will be observed for four weeks as part of Sector 26 starting around June 9 – just nine days before the predicted transit. And given the high +71.5° ecliptic latitude of KOI-4878, it will likely be observed earlier as well as part of Sector 25. It all depends on the details of the camera pointing to be chosen for these sectors over the months to come.

But before we get our hopes up too much about these TESS observations, the detection of a transit of KOI-4878.01 will be difficult to say the least. The best expected 1-hour Combined Differential Photometric Precision for KOI-4878 with a TESS magnitude of 11.8 is predicted to be on the order of about 600 ppm – about six times larger than the KOI-4878.01 transit depth observed by Kepler. According to the TESS web site, scattered light from the Earth and Moon is also expected to be a problem in Sectors 25 and 26 and, to a lesser extent, in Sector 24. This will only further complicate the detection of a transit using TESS.

While a transit of KOI-4878.01 is unlikely to trigger a hit in the TESS program’s automated data reduction pipeline, it might be detectable through the application of more vigorous statistical analysis. This endeavor will be aided by the 13-hour length of the transit event which is several times longer than typical for more commonly observed transiting exoplanets in smaller orbits around smaller stars. The longer transit will result in more data points to help improve the statistics. Even a statistically marginal detection of a transit could help build a case to confirm KOI-4878.01, if the data quality is as good as possible, if the Sun and Moon do not interfere with observations and so on.

An independent detection method like precision radial velocity measurements might eventually help. While the older generation of instruments are incapable of making measurements of the needed accuracy for a V magnitude 12.4 object, newer instruments might be up to the task. ESPRESSO (Echelle Spectrograph for Rocky Exoplanet- and Stable Spectroscopic Observations), which started making scientific observations from the European Southern Observatory’s Very Large Telescope in October 2018, could detect planets as small as Neptune orbiting KOI-4878. Future instruments with greater sensitivity could eventually detect KOI-4878.01 or refute its existence. In the meantime, we will need to wait for more transits to be observed to see if Kepler managed to find at least one true Earth-twin during its highly successful mission.

Follow Drew Ex Machina on Facebook.

Related Reading

Other articles about the planetary finds of NASA’s Kepler mission and the K2 extended mission can be found on the Kepler Mission page.

General References

A. Frasca et al., “Activity indicators and stellar parameters of the Kepler targets. An application of the ROTFIT pipeline to LAMOST-Kepler stellar spectra”, Astronomy and Astrophysics, Vol. 594A, pp. 39-39, October 1, 2016

R.K. Kopparapu et al., “Habitable zones around main-sequence stars: new estimates”, The Astrophysical Journal, Vol. 765, No. 2, Article ID. 131, March 10, 2013

Ravi Kumar Kopparapu et al., “Habitable zones around main-sequence stars: dependence on planetary mass”, The Astrophysical Journal Letters, Vol. 787, No. 2, Article ID. L29, June 1, 2014

Data Validation (DV) Report for Kepler ID 11804437 Quarters 1 – 16, NASA Kepler Mission, August 16, 2013

KOI-4878, SIMBAD website (accessed September 4, 2019) [Link]

NASA Exoplanet Archive web site [Link]

TESS Observations, TESS website (accessed September 4, 2019) [Link]